In a previous genome-wide linkage scan to screen for type 2 diabetes genes in the Pima Indians, the highest LOD score was obtained around marker D1S1677 on chromosome 1q. To further refine this segment carrying the putative diabetes susceptibility locus, we are now evaluating densely spaced single nucleotide polymorphisms (SNPs) by association analysis in selected affected an unaffected subjects. Using mainly the Whitehead Institute Database, we are now focusing on a contig of ~20 yeast artificial chromosomes (YACs) spanning an interval of about 10 cM. We are systematically exploring published as well as novel SNPs, and so far the p-values for association of individual markers range between 0.01 and 0.05. Our current goal is to investigate at least one informative SNP per each YAC, and by performing the association markers as well as their haplotypes, we anticipate that this strategy will result in a defined peak of significant association that will be used to select candidate genes for mutation screening in diabetic subjects.